In 1998, 180,000 Americans will be diagnosed with lung cancer and there will be 160,000 deaths, more than from breast, prostate and colon cancer combined. Platinum-based therapy has been accepted as a standard for treatment of locally advanced (Stage III) or metastatic (Stage IV) non-small cell lung cancer (NSCLC). As with other DNA-damaging agents, platinums are most effective against tumors containing wild-type 53. Impressive early results of clinical studies combining new antimicrotubule agents with a platinum led the Southwest Oncology Group to initiate a 400 patient randomized Phase III clinical trial (SWOG-9509) comparing two classes of antimicrotubule agents, the taxanes (paclitaxel) and vinca alkaloids (vinorelbine), each combined with a platinum. Each of these antimicrotubule agents demonstrates molecular mechanisms of action in vitro which may explain the clinical results: a) both induce apoptosis through phosphoylation and inactivation of the antiapoptosis protein Bcl2 and b) both are active in tumors that contain mutant p53. However, these activities may differ between the taxanes and vinca alkaloids based on potency, specificity, or as yet undetermined factors. Additionally, taxanes show unique clinical activity in patients with platinum-refractory NSCLC particular molecular profiles of response-related genes will differ between the taxane and vinca containing arms of this study. The specific aims are to examine the mutational status of p53 and B-tubulin and correlate the data with the tumor pathways will be examined. Tumor specimens obtained pretreatment from patients entered in SWOG-9509 will be studied. Mutational status will be examined by single-strand conformation polymorphism (SSCP), microarray and yeast if warranted. Proliferating cells will be identified by immunohistochemical (IHC) staining of Mib1 and apoptotic cells by end-labeling of fragmented DNA. Expression of Bcl2, Bcl-x(l), Bax and p27 will be assayed by IHC and rations of Bcl2:Bax and Bcl-x(l);Bax established by manual and digital scoring. Molecular data will be correlated with patient response and survival, and reviewed for statistical significance. This multidisciplinary research team is particularly well-positioned to optimize use of this important patient resource to perform this hypothesis-driven, molecular-clinical correlation. This study may lead to a better selection of patients for either combination therapy, leading to improved quality of care and clinical outcome.